本文要点
每5至7例≥50岁患者的胸腹部CT中,就有1例存在中重度椎体骨折。法国15.2万例CT显示患病率24.5%,瑞典队列16.8%,澳大利亚队列24%,数字跨国一致。 漏报是最大单一瓶颈。瑞典腹部CT队列中放射科报告率仅30%;英国全国审计(6357例)显示骨折严重程度记录率26.2%,建议进一步评估的仅2.6%。即使被报告,1年内启动骨质疏松治疗的比例约10%。三重断裂:漏检→漏报→漏治。 放射科现在能做的事:将"椎体楔形变"改为"椎体骨折";用Genant分级记录严重程度;写上"建议骨质疏松评估"。中重度骨折(Genant 2至3级)的CT判读观察者间一致性良好(κ>0.7),不需要专科培训。 临床医生现在能做的事:收到含椎体骨折字样的CT报告,不要停在"已阅";首年再骨折率19%,这是启动治疗的时间节点,不是等下次复诊的理由。 AI已有多工具通过临床验证(汇总敏感性83%,特异性92%)。瑞典2026年临床实施研究将AI接入放射科工作流,4个月扫描3971例,每天新识别2至3例需要干预的患者,49%转入骨折联络服务。技术不是瓶颈,流程对接是。
本期判断
三重断裂的量化
放射科能做什么:现在,不等AI
接诊医生能做什么:报告到了之后
AI的位置:下一层解决方案
一个实际的问题
写在最后
参考文献
Roux C, Rozes A, Reizine D, et al. Fully automated opportunistic screening of vertebral fractures and osteoporosis on more than 150 000 routine computed tomography scans. Rheumatology (Oxford). 2022;61(8):3269-3278. Wiklund P, Buchebner D, Geijer M. Vertebral compression fractures at abdominal CT: underdiagnosis, undertreatment, and evaluation of an AI algorithm. J Bone Miner Res. 2024;39(8):1113-1119. Kolanu N, Silverstone EJ, Ho BH, et al. Clinical utility of computer-aided diagnosis of vertebral fractures from computed tomography images. J Bone Miner Res. 2020;35(12):2307-2312. Howlett DC, Drinkwater KJ, Mahmood N, et al. Radiology reporting of osteoporotic vertebral fragility fractures on computed tomography studies: results of a UK national audit. Eur Radiol. 2020;30(9):4713-4723. Howlett DC, Drinkwater KJ, Mahmood N, et al. Radiology reporting of incidental osteoporotic vertebral fragility fractures present on CT studies: results of UK national re-audit. Clin Radiol. 2023;78(12):e1041-e1047. Delmas PD, van de Langerijt L, Watts NB, et al. Underdiagnosis of vertebral fractures is a worldwide problem: the IMPACT study. J Bone Miner Res. 2005;20(4):557-563. Buckens CF, de Jong PA, Mol C, et al. Intra and interobserver reliability and agreement of semiquantitative vertebral fracture assessment on chest computed tomography. PLoS One. 2013;8(8):e71204. Gete KY, Idris ZN, Ayele BA. Deep learning for opportunistic vertebral fracture detection on routine thoraco-abdominal computed tomography: a systematic review and hierarchical summary receiver operating characteristic meta-analysis. Acad Radiol. 2026;S1076-6332(26)00421-6. Woisetschläger M, Baldimtsi E, Lindblom M, et al. Clinical implementation of AI for vertebral fracture detection in CT aligned with fracture liaison services: high prevalence of undiagnosed vertebral fractures. Osteoporos Int. 2026. doi:10.1007/s00198-026-07907-9 . O'Dowd EL, Tietzova I, Bartlett E, et al. ERS/ESTS/ESTRO/ESR/ESTI/EFOMP statement on management of incidental findings from low dose CT screening for lung cancer. Eur J Cardiothorac Surg. 2023;64(4):ezad302. Diacinti D, Vitali C, Gussoni G, et al. Misdiagnosis of vertebral fractures on local radiographic readings of the multicentre POINT study. Bone. 2017;101:230-235. Ensrud KE, Schousboe JT. Vertebral fractures. N Engl J Med. 2011;364(17):1634-1642.
名词缩写表
缩写 | 中文全称 |
CT | 计算机断层扫描(Computed Tomography) |
LDCT | 低剂量CT(Low-Dose Computed Tomography) |
DECT | 双能CT(Dual-Energy Computed Tomography) |
MRI | 磁共振成像(Magnetic Resonance Imaging) |
STIR | 短时反转恢复序列(Short Tau Inversion Recovery) |
AI | 人工智能(Artificial Intelligence) |
FLS | 骨折联络服务(Fracture Liaison Service) |
PACS | 影像归档与通信系统(Picture Archiving and Communication System) |
HU | 亨氏单位(Hounsfield Unit) |
DXA | 双能X线骨密度仪(Dual-energy X-ray Absorptiometry) |
BMD | 骨密度(Bone Mineral Density) |
RCR | 英国皇家放射学会(Royal College of Radiologists) |
NNS | 需要筛查数(Number Needed to Screen) |
夜雨聆风